Thursday, October 1, 2015

CASAA FDA comment on UCSF FDA comment - How Public Health Policy May Be Hazardous to Your Health




To: U. S. Food and Drug Administration

From: Brian L. Carter, PhD and Carl V Phillips, PhD

30 September 2015

VIA REGULATIONS.GOV
Comment Tracking Number: 1jz-8lf9-1nx4

Re: Comments on ANPRM “Warnings and Child-Resistant Packaging for Liquid Nicotine, Nicotine-Containing E-Liquid(s), and Other Tobacco Products” (Docket No. FDA-2015-N-1514)

This is a comment on Docket No. FDA-2015-N-1514 (advance notice of proposed rulemaking (ANPRM), “Nicotine Exposure Warnings and Child-Resistant Packaging for Liquid Nicotine, Nicotine-Containing E-Liquid(s), and Other Tobacco Products”). We are concerned scientists with expertise in fields related to the proposed rule making. Carter is a member of the Board of Advisors of CASAA (The Consumer Advocates for Smoke-free Alternatives Association) and an independent consultant. Phillips is Chief Scientific Officer of CASAA.

This comment should not be mistaken for CASAA’s direct comment on the ANPRM [posted at regulations.gov at http://www.regulations.gov/#!documentDetail;D=FDA-2015-N-1514-0385 and also appearing at http://blog.casaa.org/2015/09/casaa-comment-on-fdas-proposed.html] (hereafter referred to as “the CASAA comment”). That comment presented analysis that supported the inclusion of safety warning labels and the use of child-resistant packaging, so long as these were not inappropriately used to mislead consumers about the risks or to intentionally lower product quality. The present comment has been endorsed by the CASAA leadership, and thus should also be considered an second official comment by the organization.
In the first CASAA comment, we predicted that those who seek to discourage e-cigarette use and other tobacco products may use this as an opportunity to misuse the legitimate process of creating valid and potentially beneficial safety warnings to deliver misleading claims with the intent of lowering the appeal of these products for consumers. The present comment is intended to make that concrete by observing exactly this behavior in one specific comment that has been filed. We reiterate the observation made in the CASAA comment that if FDA were to follow such advice, it would be scientifically misleading and grossly inappropriate public policy, would increase the risk of accidental poisonings by causing consumers to dismiss the warnings on these and other products as unreliable, and unnecessarily discourage smokers from seeking low-risk alternatives.
We are taking as an example for our response the comments submitted to the FDA under this docket number by the University of California, San Francisco, Tobacco Center of Regulatory Science (TCORS) and the California Poison Control System (CPCS) (hereafter: the TCORS/CPCS comment). This comment was previously published by Professor Stanton Glantz, PhD, on his UCSF website
[http://www.tobacco.ucsf.edu/ucsf-and-california-poison-control-centers-response-fda-questions-liquid-nicotine-products]
and appears at regulations.gov at
[http://www.regulations.gov/#!documentDetail;D=FDA-2015-N-1514-0018.]

We expect that this is but one example of many comments from commentators with similar ideology-based political goals willing to subvert the rulemaking process in support of those goals. This is not an attempt to identify every error or weakness in the particular comment, but to illustrate the attempts to misuse this process to inappropriately demonize the products. We expect that this pattern will continue with every proposed regulation in this space and point out that FDA should not mistake political advocacy for scientific analysis, even when it is coming from the NIH- or FDA-funded academics. We wish to emphasize the contrast between the tendentious TCORS/CPCS comment and the CASAA comment, which we believe makes appropriate, reasonable, and science-based recommendations concerning the labeling and packaging of e-cigarette components and dissolvable smokeless tobacco products. The authors of the TCORS/CPCS comment are on record as opposing the use of low-risk tobacco products and have supported campaigns to dissuade consumers, including smokers, from experimenting with them (e.g., http://sanfranciscotobaccofreeproject.org/curbit-2/, https://tobacco.ucsf.edu/san-francisco-launches-e-cig-public-education-campaign).

Text from the published TCORS/CPCS document is presented throughout, indented and in italics, followed by our comments in a point-by-point analysis. The TCORS/CPCS comments are reproduced verbatim; the grammar and typographical errors appear in the original.

“Liquid nicotine and nicotine-containing e-liquid(s) (liquid nicotine combined with colorings, flavorings, and/or potentially other ingredients)” are generally referred to as liquid nicotine. (Page 1)
Although “liquid nicotine” is a term used by the FDA in the ANPRM, a scientist should immediately recognize it as inaccurate when applied to the products under consideration. First, the term is redundant. Nicotine, at standard temperature and pressure, is a liquid. Second, using this term implies the solutions used by e-cigarettes consist of pure nicotine, which would be enormously more hazardous than the actual products. “Liquid nicotine”--i.e., pure nicotine--has only industrial uses, including the manufacture of e-cigarette refill liquids, and its handling and labeling are governed by workplace safety regulations. In contrast, the nicotine solutions used by the general public are dilute solutions containing as little as .3% nicotine, and very seldom exceeding 3.6%.

Calls to Poison Center in the U.S. have been increasing steadily since 2010, with calls per month increasing from one in September 2010 to 512 in February 2014. Exposures were most commonly ingestions (68.9%), followed by inhalations (16.8%), eye exposures (8.5%) and skin exposures (5.9%). E-cigarette calls involves children age 5 or less (51.1%), with 42.0% calls from peole > 20years. E-cigarette calls were more likely to report and adverse effect (57.6%) coampared to cigarette ingestions (36.0%). The most common adverse effects were vomting, nausea and eye irritation, with one suicide death from intravenous injection. (Page 2)

First, this is factually incorrect; as noted in the CASAA comment, the incidence of calls declined in the most recent period reported, rather than “increasing steadily” [http://www.fda.gov/downloads/TobaccoProducts/NewsEvents/UCM454403.pdf, p.520; https://www.youtube.com/watch?v=rQSRG6hmh-c, at and around 2:02:00]. Moreover, that statement is in absolute terms; the ratio of calls to usage has decreased greatly over the years. Clearly, the handling of these products by consumers is becoming safer over time. More important, as predicted in the CASAA comment, the increase in these calls over a period starting before e-cigarettes became popular is being used to imply that there is an important problem. The low absolute magnitude of these calls compared to those for other household products is ignored. As noted in the CASAA comment, such a relative increase is inevitable. Moreover, such calls are not a measure of hazard, but of alarm on the part of consumers, much of which has likely been manufactured by misleading messaging.  There have been countless popular press stories, engineered by those who vilify e-cigarettes, which would tend to create unwarranted alarm [e.g., http://www.abcnews4.com/story/29696942/liquid-nicotine-from-e-cigs-poses-poison-danger-to-kids].   

Given these findings, the FDA should require package inserts for liquid nicotine similar to those provided with prescription drugs that describe their hazards of liquid nicotine. (Page 2)

This, which followed immediately after the previous quotation, is a non sequitur. The stated reason (merely the above observation that poison control calls have increased) offers no justification for this or any other specific intervention. The TCORS/CPCS comment offers no affirmative justification for a pharmaceutical insert intervention. Such inserts are designed to provide extremely detailed information, while the present goal is to provide easily accessible information that will be noticed and quickly understood.

Warning labels should include the dermal absorption hazard, a statement that the amount of nicotine in e-liquids can be substantial and pose a serious risk, especially for small children, and that care should be taken to keep products out of the reach of children and pets. (Page 2)

Providing advice about dermal exposure is a sensible regulation, as noted in the CASAA comment, provided this information is concise and does not overstate this relatively minor risk. However, the TCORS/CPCS comment dramatically overstates, and offers no evidence, that dermal exposure is a “serious” risk. Similarly, the CASAA comment notes the appropriateness and value of a “keep out of reach of children and pets” warning.

Here are the relevant sections of the information from safety data sheets required of chemical suppliers: (Page 2)

2.2 GHS [Globally Harmonized System] Label elements, including precautionary statements
Screen Shot 2015-08-15 at 2.20.15 AM.png
Signal word Danger
Hazard statement(s)
H300 + H310 Fatal if swallowed or in contact with skin
H410 Very toxic to aquatic life with long lasting effects.
Precautionary statement(s)
P262 Do not get in eyes, on skin, or on clothing.
P264 Wash skin thoroughly after handling.
P270 Do not eat, drink or smoke when using this product.
P273 Avoid release to the environment.
P280 Wear protective gloves/ protective clothing.
P301+P310 IF SWALLOWED: Immediately call a POISON CENTER or doctor/physician.
P302+P350 IF ON SKIN: Gently wash with plenty of soap and water.
P310 Immediately call a POISON CENTER or doctor/ physician.
P361 Remove/Take off immediately all contaminated clothing.
P363 Wash contaminated clothing before reuse.
P391 Collect spillage.
P405 Store locked up.
P501 Dispose of contents/ container to an approved waste disposal plant.

This is the Safety Data Sheet for pure (99+%) nicotine, and as noted, it applies to “chemical suppliers.” It applies to none of the products covered under this ANPRM, and is therefore irrelevant. Its inclusion in the TCORS/CPCS comment is either evidence of ignorance on the part of the authors, or worse, a deliberate attempt to mislead the FDA by grossly exaggerating the toxicity of the products under review. In either case, it represents prima facie evidence that the TCORS/CPCS authors are acting in bad faith; they are either claiming expertise that they lack, or are engaging in deceptive tactics to pursue an agenda that is contrary to the purpose of the ANPRM and advocating policies that are antithetical to the advancement of public safety. Following the logic of these authors, the similarly dire Safety Data Sheet warnings about glacial acetic acid should appear on bottles of vinegar.

This is the first of many places where the TCORS/CPCS authors seem to not recognize the difference in toxicity between pure nicotine and the small quantities of nicotine found in the consumer products addressed by the ANPRM. It is a standard practice in tobacco control and public health (as well as anti-vaccine activism, anti-GMO activism, and countless other anti-scientific efforts) to mislead consumers speciously into believing that a substance that contains any of a particular molecule has the same properties of a large quantity or high concentration of those molecules. But it is notable in this case that the authors would not just be seeking to mislead the consumer with the phrases suggested below; since this is appearing in a comment to the FDA, they are trying to mislead FDA scientists into making this same error.

...there is evidence to suggest damaging oral health effects from chronic exposure to the nicotine even when nicotine-containing products are used as intended. Tobacco smoking is the most critical modifiable risk factor in the development of periodontal (gum) disease, (Page 6)

This is the first of many instances where the TCORS/CPCS authors improperly cite research related to smoking with the clear intent of drawing a false equivalency between that exposure and exposure to nicotine (either intentional or accidental), and make clear that the true goal of their proposals have little to do with advancing the goals stated in the ANPRM. There is simply no reason to believe that the effects of smoking apply to the products under consideration in the ANPRM. Even if such a risk did exist, this observation in the TCORS/CPCS comment would be completely irrelevant to the present ANPRM, since this would relate to the intended use of the products, not accidental exposure hazards.

Nicotine itself may contribute to the development and persistence of periodontal disease, separate from the potentially damaging effects of other toxicants found in tobacco smoke. This major impact of smoking on poor gum health has been mechanistically implicated to the nicotine contained in cigarette smoke. Nicotine impairs the ability of human gingival fibroblasts to adhere to tooth root surfaces, which is essential for inflamed periodontal tissues to heal after therapy, and nicotine may suppress the cytodifferentiation and mineralization capacity of periodontal ligament cells. Therefore, any nicotine-containing product has the potential to damage oral tissues, and the FDA should consider oral health related warnings for all nicotine-containing products. (Page 6)

It is notable the TCORS/CPCS comment goes into detail about this tangent (which is completely irrelevant to the question of safety warnings and packaging) while basing their relevant claims on single-sentence assertions. Since this is irrelevant to the matter at hand, we will not bother to rebut the specific claims other than to note that the authors do not seem to understand that finding a claim in a single piece of obscure, esoteric, and preliminary research does not constitute scientific analysis, nor do these obscure findings constitute a significant body of knowledge that justifies regulatory action.

Warnings on liquids should contain specific information, such as the concentration of nicotine in the liquid and the amount of liquid in commonly understood measures (like teaspoons) as well as in that the more technical measures(mL) that could poison or kill an infant. For example - a recent estimate of the oral LD50 for nicotine is 6.5-13 mg/kg. For a 15 kg child a potentially lethal dose is less than 100 mg. When the lethal dose is stated on the label, users of these liquids may take the warnings more seriously since they can visualize how much liquid constitutes a potentially lethal dose. (Page 7)

This suggestion is clearly at odds with labeling and standard best practices for consumer products with similar toxicity. There is no evidence offered, citation provided, or even argument asserted about why such a significant departure from standard warning label practice is warranted. Moreover, the stated quantification suggests much higher risk than the best evidence-based estimate of the lethal dose, as noted in the CASAA comment.

This suggestion is ill considered and dangerous for a number of reasons: 1) There is no evidence that displaying lethal doses on products is helpful to consumers or leads them to take such a warning more seriously. 2) The vast majority of consumers are non-scientists and do not realize that a figure derived from an LD50 is an estimate of potential lethality that has only approximate relevance when applied to individual cases. 3) This estimate is likely to be poorly understood by consumers and stands a strong likelihood of creating either panic and reckless actions in the case of accidental exposure or false reassurance in cases of exposure that fall short of the noted quantities. 4) This information is routinely not included on other products because this information can be used as the basis of murder or suicide attempts. 5) Most users of these products will immediately recognize this as an attempt to introduce anti-nicotine propaganda into the labeling and thus may be inclined to ignore the warning label entirely.

It is apparent that the authors are attempting to use accidental exposure warnings to further their personal political goals of vilifying nicotine-containing products, even at the expense of making those warnings less effective and potentially dangerous.

Given how new electronic cigarette products are, and how little information is available to the public regarding the health effects of e-cigarettes (and other newly emerging products) on people in general or vulnerable populations, it is critical to include as much information as possible.
At the same time, the FDA should be cognizant of the fact that if you provide too much information people will ignore it (for example full drug labels handed out with medications). (Page 7)

The assertion that there is “little” information available to the public regarding the health effects of e-cigarettes is simply incorrect, as a Google search will demonstrate. (Indeed, most of the information contained in such a search will yield incorrect information that overstates the risks.) As for the notion that it is “critical” to include as much information as possible, this is antithetical to the practice of effective warning labeling, which the authors themselves note in the flatly contradictory sentence that follows. The authors, given the total amount of information they are suggesting be included, are advocating warnings totalling hundreds of words, which they explicitly acknowledge will likely be ignored.  

We suggest the following warning statements for nicotine-containing products (related specifically to nicotine; this list does not include other items related to the consumption of tobacco products): (Page 8)

The warning statements below are clearly reflective of the considerable bias of the authors and their willingness to ignore or even undermine the goals of the ANPRM and mislead for the purpose of manipulating consumer choices.
  • This product can cause and maintain nicotine addiction

Even if this claim had strong empirical support, which it does not, it would obviously be unrelated to the risk and effects of an accidental exposure event. This is sufficient to show that the TCORS/CPCS authors are given to making specious and tangential points that lie outside the scope of the ANPRM, and are willing to do so without scientific support.

This statement sheds further light on the authors’ inattention to scientific detail. What is known about the relationship between nicotine use and addiction is almost exclusively derived from research with smokers. What little is known about non-combustible nicotine use suggests weak empirical support in terms of addiction (Frenk & Dar, 2000; Dar & Frenk, 2011). For example, research on long-term use of nicotine replacement products indicates only a slim minority of these users find them “addictive” enough to use indefinitely. A number of scientists have noted the considerable conceptual confusion regarding a workable scientific definition of addiction (e.g., Albanese & Shaffer, 2012). However, whatever “addiction” means, there is no evidence that it definitively describes the use of the products in question, particularly e-cigarettes.


  • Nicotine is a poison: Do not drink the liquid in this product

The use of the sentence “Nicotine is a poison” is unjustified and misleading in all of the warnings on this list. The authors are presumably trying to imply that warning effectiveness is increased by explaining why someone should not consume a product in a particular manner, though they do not even say this, let alone attempt to justify it. It is unsupported by evidence and thus is not done for other warning labels. It apparently serves as a pretext for raising the level of alarm in consumers to dissuade them from using the product as intended. It would be equally technically accurate and equally harmful to honest communication and consumer understanding to say “Acetaminophen is a poison” (and, indeed, it poses greater poisoning risk than e-cigarette liquid).

There is no reason to believe that a “Do not drink” label on a product that every consumer knows is not for drinking would serve any useful purpose. Indeed, it seems like most likely to convey the message that all of the warning label content is equally insipid and only meant for people who are completely ignorant of the intended use of the product.

  • Nicotine is a poison: Do not let the liquid touch your skin

This warning is such an absurdly strong overstatement that it constitutes out-and-out disinformation. Moreover, it is impossible to comply with since most every user of nicotine refill liquids gets small amounts of the liquid on their skin. This is easily remedied by wiping it off, but even without that remedy, such exposures simply deliver a dose of nicotine that is in the same order as from intentional use of the products.

As noted in the CASAA comment, consumers are not necessarily aware of the fact that nicotine (and many other chemicals) can be absorbed through the skin, so an appropriate warning that communicates this could be useful in the event of a major dermal exposure event for a small child. However, the proposed label would inevitably generate derision by those who know from experience that letting the liquid touch their skin causes no problem, undermining accurate communication of this rare, but real, risk. Once again, the TCORS/CPCS authors show a willingness to sacrifice the legitimate goals of the ANPRM in pursuit of misleading communication.
  • Nicotine is a poison, even in very small amounts.

This phrasing is vague to the point of uselessness, since it depends of what “small amounts” means to the reader. A warning that communicates great caution about even seemingly minor exposures (one that is written properly, unlike this one) is warranted for a few extremely toxic substances (which, for obvious reasons, are not household products). But in this case the claim is simply false. The quantity of these products required to pose a toxicity hazard is substantially larger than a reasonable interpretation of “very small amounts.” It appears that the authors may be trying to mislead intentionally , addressed above, by making a statement about pure nicotine and then trying to imply it is relevant to any product that contains any nicotine.  
  • Nicotine is a poison: less than a teaspoon can kill you.

This is clearly an example of trying to mislead the consumer (and the FDA) into confusing the effects of pure nicotine with the hazard from the consumer products. There is little doubt that the consumer would read this as saying that less than a teaspoon of the product can be fatal. Since consumers would know from experience this is false, this would tend to undermine the credibility of the entire warning.
  • To prevent poisoning keep this product away from babies and children and pets

This warning is needlessly wordy. “Keep out of reach of children and pets” is sufficient and more precise. The inclusion of the word “babies” appears to be an attempt to play on the fears of parents since babies are a subset of children. There is no evidence offered that babies (implicitly, children that are too young to open a bottle) are at particular risk from these products to the point their specific mention is required in a warning statement.

  • Wear gloves when handling liquid nicotine
  • Wear protective gloves when refilling your e-cigarettes

As previously noted, this is absurd; such advice is warranted only for the high concentrations that are used in manufacturing, not the consumer products that are the subject of this ANPRM. Moreover, as far as we are aware, none of the millions of users of these products believe gloves are necessary when handling these products. This is another example of a “warning” that would be so patently absurd that it would undermine the credibility of the legitimate advice.
  • Pregnant women should not consume nicotine in any form due to its harmful effects on the developing fetus.

This is another example of the TCORS/CPCS authors apparently not understanding the goal of this ANPRM. If the authors had any evidence that accidental exposure to these products posed a notable risk for fetuses, perhaps such a warning would be relevant to the ANPRM. But since accidental exposure by an adult is almost certainly limited to those who are consuming the product intentionally, and the quantities would be similar to those being intentionally consumed, any such risk is irrelevant for present purposes.

In total, the warnings suggested by the authors above are plagued by a lack of understanding of the purpose of accidental exposure warnings, a lack of understanding of the difference between industrial and consumer concentrations of nicotine, a lack of understanding of the way in which these products are used and understood by their consumers, and a lack of understanding of even the most basic principles of effective warning design.

In fact, there is sufficient evidence to conclude that nicotine exposure increases the risk for other adverse health effects for all individuals, including otherwise healthy adults. (Page 8)

This appears to have no relevance to the matter at hand, though it is difficult to determine because it  is asserted without elaboration, let alone any reference to research. Therefore, it should be ignored as more irrelevant politically motivated speculation on the part of the authors.

In addition to warning labels on the outside of bottles and other packages, there should be inserts that list all major hazards and precautions. Based on our answer to Question A1, these inserts should include the following information:

2.2 GHS Label elements, including precautionary statements
Screen Shot 2015-08-15 at 2.20.15 AM.png
Signal word Danger
Hazard statement(s):
Can be fatal if swallowed or in contact with skin
Very toxic to aquatic life with long lasting effects.
Precautionary statement(s):
Do not get in eyes, on skin, or on clothing.
Wash skin thoroughly after handling.
Do not eat, drink or smoke when transferring this product.
Avoid release to the environment.*
Wear protective gloves
Wear protective clothing if handling large quantities*
IF SWALLOWED: Immediately call a POISON CENTER or doctor/physician. IF ON SKIN: Gently wash with plenty of soap and water.
Immediately call a POISON CENTER or doctor/ physician.
Remove/Take off immediately all contaminated clothing.
Wash contaminated clothing before reuse.
Collect spillage.*
Store locked up.*
Dispose of contents/ container to an approved waste disposal plant.*

*Most appropriate for containers of large quantities handled by wholesalers and retailers, i.e. greater than the one ounce or so bottles generally sold to consumers. (Page 11)

With this suggestion, that all nicotine-containing products--even those that contain a dilute solution of 0.3% nicotine--should include an insert of the Safety Data Sheet for 99+% nicotine, the authors have committed a bright-line ethical violation, especially because they have published this contention for the general public as noted above. It is clearly tendentious and dishonest to suggest this is appropriate or useful. Moreover, including this insert in all nicotine-containing products is likely to produce considerable harm to the public. 1) This information is incorrect and does not apply to the products it will be associated with. Deliberately misleading consumers completely undermines all legitimate government provided health information. 2) Harm will be inflicted on consumers by unnecessarily wasting their time with needless precautions at the least, and at the worst will undoubtedly cause some to engage in panicked and dangerous actions in case of trivial accidental exposure (e.g., recklessly driving to emergency rooms or needlessly inducing vomiting). This suggestion by TCORS/CPCS is irresponsible.

The FDA should require nicotine exposure warnings be placed on product packaging and that prominent, strong warnings be placed at the physical locations where nicotine-containing products are sold, including point-of-sales as well as on any websites of online retailers. (Page 12)

The authors compound the problem by suggesting the FDA require Safety Data Sheet statements be displayed at retail locations and websites selling these products. They claim that

Prominent warnings in retail locations have the potential to educate the public broadly and to inform consumers’ decision-making before they have committed to purchasing a nicotine- containing product. (Page 12)

The suggestion that the public in general, most of whom are not interested in these products, nor have any intention of using them, should be broadly “educated” about them with point-of-sale “information” displays is indefensible on its face. The idea that point of sale warnings of the content above will “inform consumers’ decision-making” before purchasing is, at best, disingenuous. The clear intent appears to be to further mislead, frighten, and dissuade consumers from purchasing these products and influencing public opinion against their use.

In addition, earlier studies conducted at UCSF qualitatively and quantitatively evaluated a print ad designed to discourage smokers from starting to use smokeless tobacco. The results of this research are relevant to the question FDA is posing here. (Page 12)

Here the authors have again dramatically departed from the clear intent of the ANPRM. Communication designed to manipulate consumers into not purchasing and using the products is obviously irrelevant to safety warnings and packaging.

Though this comment is irrelevant to the ANPRM, it is worth noting that the TCORS/CPCS authors are suggesting that FDA ignore its responsibility to provide factual, science-based information to consumers concerning the safe use of these products for those who choose to use them. It is not the mission nor the legally mandated role of the FDA to use their regulatory power to unduly influence personal consumer decisions to suit policy preferences.

Graphic elements such as pictures and imagery may be more effective than text-only messages at communicating health risks of tobacco products.  Studies have found that users and non-users rate warnings with graphic images as more effective than text-only warnings. A 2015 meta- analysis of experiments and international literature examining the impact of pictorial cigarette pack warnings found that pictorial warnings are more effective than text-only warnings. (see The WHO Framework Convention on Tobacco Control (FCTC) and the FCTC Article 11.) In particular, compared with text-only warnings, pictorial warnings attracted and held attention better. (Page 17)

Once again, the authors seem to not understand the scope of the this ANPRM, since they fail to suggest how these observations are relevant to safety warnings and packaging. It is difficult to see how claims about what methods better discourage consumer choice- claims which also happen to be rather tenuous--inform us about what methods are best for encouraging safe handling of the product. FDA is presumably aware of the large body of research on making safety warnings most effective, and is aware that the observations in that paragraph are irrelevant to it.

All packages should require prominent, easy-to-read and understand text and graphical warnings. If the nicotine container is small, it can be packaged attached to a larger piece of cardboard or other packaging that is large enough to communicate the warning effectively. Such packaging is common for small items (often to deter theft). (Page 19)

The authors make an affirmative claim about imposing a costly and burdensome detail in the proposed regulation without so much as suggesting it has benefits. There is not even an assertion --let alone substantive justification--that such a requirement would improve the safety of these products. There are thousands of common consumer products on the market with similar toxicity that do not require such packaging with large warning statements. Anyone familiar with consumer behavior would realize that such extra packaging will be removed and discarded without a glance, making it difficult to believe there is any value in it. Thus anyone making such an assertion needs to justify it. Moreover, as previously stated, a package festooned with warnings and statements, such that they take up the vast majority of the package surface, are likely to be ignored by consumers as excessive and unnecessary, undermining the credibility of the content, including the statements that are useful and appropriate.    

Warning labels should not vary or be altered because of differences in packages or even stated differences in nicotine levels. For example, numerous studies show that there is a great deal of variation between the stated amount of nicotine in these products and the actual amount, with studies showing instances of e-cigarette products being labeled as zero nicotine and actually containing nicotine. (Page 19)

This statement is a combination of hyperbole and fallacious reasoning. If there are “numerous” studies showing a variation in nicotine content among products the authors should cite them. However, none are cited for this claim. The claim that there is “a great deal of variation” between actual and stated nicotine content in these products also has no citation. To our knowledge, in the rare reported cases where deviation from the labeling has been detected, the discrepancy was far too small to substantially affect the risks from accidental exposure. There is no evidence, or even an allegation that we are aware of, that a single person has been harmed by minor variations in nicotine content. The citation used in this paragraph refers to a single study of mislabeled non-nicotine-containing liquid. That is, nicotine was found in a subset of products of a single brand of non-nicotine fluid out of thousands on the market. This finding, although almost certainly an anomaly, may argue for stricter quality control measures placed on the manufacturers of these products, but it does not offer any support or guidance for how nicotine-containing products should be labeled.

Given the risk to children from the accidental ingestion of these products, the warnings should remain on the label even when the vial is child resistant since by definition, it is only resistant not child-proof and the dangers are significant. There could be cases where child resistance fails. (Page 20)

It is difficult to understand what the authors are arguing here. The authors seem to imply that an exception might be granted for mandated safety warnings when a package is child resistant. We are aware of no regulation that allows this, or anyone who seriously argues it is wise to allow it.

A dog who ate an e-cigarette cartridge died as a result. The Pet Poison Helpline has encountered a sharp uptick in calls concerning cases of nicotine poisoning in pets that ingested e- cigarettes or liquid nicotine refill solution. In six months in 2014, cases more than doubled, indicating that along with their increased popularity, e-cigarettes are becoming a more significant threat to pets. While dogs account for the majority of cases, nicotine in e-cigarettes and liquid refill solution is toxic to cats as well. (Page 20)

Protection of pets is a legitimate concern for safety regulations and procedures. However, this comment offers no guidance on the specific issue of how warning labels should be designed to protect pets. Because any such warnings would inevitably refer to “pets” rather than a particular species, the speculation about cats seems particularly uninformative. It is also worth noting that the the cited statistic--a relative increase, from a single data series, for an arbitrary and unjustified historical period--should leave the reader strongly suspicious that the pet website authors used an unusually low month in the available data to use as a baseline in order to present a relative increase that was unrepresentative of the overall trend.

The FDA should not limit any proposed warnings to text-only messages that are small in size. A systematic review of 94 qualitative and quantitative research studies concluded that small, text-only messages are less effective at conveying information to consumers than large warning with pictures. (Page 20)

The review cited in this passage concerns health warnings to product users placed on cigarette packs. Once again, the authors are (a) failing to understand that the ANPRM is about accidental exposure warnings, not user health warnings and (b) inappropriately using cigarette related research as if it should apply to e-cigarettes and other novel tobacco products. That is, the warnings in that review article were clearly intended to discourage use of the products, and therefore do not apply to the safety-related goals of the current ANPRM.

The authors are suggesting FDA consider the use of pictures in warning labels. They do not provide any examples of what pictures should be included, but it is difficult to imagine how the use of pictures of the type used in the cited review could possibly serve a legitimate purpose for conveying product safety information.

While we are not aware of work that examines nicotine exposure warnings directly, the general consensus is the bigger the warning the better, and visual elements are more impactful (visual salience, knowledge, and behavioral intentions) than text warnings.

Once again, the TCORS/CPCS authors are citing research (here, and in subsequent paragraphs) related to public health efforts to dissuade consumers from smoking cigarettes. The graphic elements under review in this research are clearly irrelevant to the purposes of this ANPRM.

14. Are there any nicotine exposure warnings (textual and/or graphic) for liquid nicotine required by authorities at the local, State, or Federal (i.e., other agencies) level, or by foreign governments that you particularly would like to highlight? If so, which ones and why? Are there any data regarding the effectiveness or utility of these warnings? If so, please provide these data.
California requires a warning about the reproductive risks of nicotine under its Proposition 65, but these warnings are not particularly effective. Businesses decide whether to place a warning label based upon their understanding of the chemicals in the product. Additionally, the manufacturer is not required to provide the California Office of Health Hazzard Assessement with information about the products and the labels only indicate there’s a chemical that might cause cancer or affect reproduction. There is no additional information for the consumer.

Guernsey (a Channel Island) tested a warning on nicotine products but it was not implemented. Suffolk County, NY requires liquid nicotine warning notices posted at cash registers where the product is sold but labels are not required on bottles. In 2014, Suffolk County proposed a bill that would have required the listing of ingredients for nicotine products. A few bills have been introduced by city and state legislative bodies around warning labels on products and Farmington Connecticut introduced a bill requiring child-proof containers for liquid nicotine. (Page 22)

This passage is unresponsive to the questions above. It is unclear how California Proposition 65 warnings, unimplemented warning measures of a small foreign community, and proposed warning bills from a New York county could constitute useful information for the FDA. The authors offer no data that speaks to the effectiveness of the point of sale warning in Suffolk county, and in the case of California proposition 65 warnings, they concede they are largely ineffective.
B. Child-Resistant Packaging

1. Should FDA require child-resistant packaging for liquid nicotine? If so, why?
Yes. As noted above, there is a growing number of accidental poisoning cases among children. In addition, it is likely that young children may find little vials of “kid friendly” flavorings enticing enough to try. (Page 23)

As noted in the CASAA comment, we favor standardized child-resistant packaging for all nicotine containing liquids. That children may be more attracted to certain flavoring agents in these liquids is both purely speculative and irrelevant. The phrase “kid friendly,” which has no scientific meaning, nor even a clear common-language meaning, is often used by these authors as a polemic against these products, further illustrating their orientation as political actors willing to forgo science-based regulations in favor of policy-based goals.

2. Should FDA require child-resistant packaging for liquid nicotine if the liquid nicotine product is not intended to be opened by the consumer (e.g., liquid nicotine in permanently sealed, prefilled, and/or disposable cartridges)?

All liquid nicotine products (and, in fact, all nicotine products) should be in child-resistant containers. The intent of the manufacturer is not the question; it is the ability of children to consume the product in a way that can lead to poisoning. (Page 23)

We agree that warnings and child-proof packaging should be based on the “ability of children to consume the product in a way that might lead to poisoning.” However, the TCORS/CPCS authors do not seem to really believe this themselves. Their suggestion about “all nicotine products” would include cigarettes and even premium cigars, which pose no substantial risk of serious poisoning. Permanently sealed disposable e-cigarette cartridges present very small risk for accidental poisoning by consumption. These small cartridges each contain a quantity of nicotine that is far below a toxic dose. Moreover, these are already “child-resistant containers,” being difficult to open without tools and considerable effort. However, the TCORS/CPCS authors are apparently so opposed to recreational nicotine use that they fail to recognize that these cartridges do pose an accidental choking hazard, which we address in our CASAA comment; protecting against this requires child-resistant packaging rather than containers.
8. Are there other factors FDA should consider to further prevent or discourage people (especially infants and children) from inadvertently consuming or being exposed to liquid nicotine? If so, please explain. Examples of other factors may include: attractiveness of the product or packaging (e.g., appealing images, fragrance, flavors), resemblance of packaging to food and drink items (e.g., candy, fruit), color of the product (e.g., resemblance to beverages such as juice), resemblance of packaging to that of medications (e.g., eye drops).
Since the attractive, kid friendly labeling can be alluring to children, we recommend that vials of nicotine should be plain black and white with information identifying percent or mg nicotine and name of flavor but without pictures of said flavors or colors. Flavors should be assigned alphanumeric codes, such as Z34, instead of descriptive flavor names that may appeal to children, such as cherry. (Page 24)

The authors fail to offer any evidence that suggests such an action would change the risk of accidental poisoning. Moreover, they make no attempt to define “kid friendly labeling,” let alone how it differs from ubiquitous packaging features that are are designed to appeal to adults. Countless household products of similar toxicity contain pictures and names of fruit (e.g., lemon-scented cleaning agents; cherry-flavored OTC medicines), including FDA-regulated nicotine-containing products (e.g., Nicorette “Fruit Chill” lozenges). It is difficult to see why the labeling of the products covered under this ANPRM constitute an extraordinary case, where deviation from standard labeling practices of consumer products with similar toxicity, is warranted. The TCORS/CPCS authors do not so much as suggest why this might be so.

If the authors could have produced any evidence that suggests requiring plain black and white labeling and alphanumeric codes might reduce the risk of accidental poisoning they presumably would have done so. In what appears above they repeatedly invoke scientifically tenuous claims that are tangentially related to the questions at hand. Thus this suggestion appears to be nothing more than an attempt to use the ANPRM and its legitimate goals of child safety as an excuse to gratuitously lower the quality and appeal of these products for adults. If these authors were following the serious intent of the ANPRM by providing carefully considered recommendations, they would not have suggested this radical policy without some evidence-based justification for it. If they had offered such justification, then as responsible commentators they would have addressed the costs of such an action--at the very least in terms of its constitutional and restraint of trade implications and the lawsuits against FDA that would likely follow as a result.

This failure illustrates why these authors--and the others who wrote comments that tried to use this ANPRM as an excuse for lowering product quality--should not be taken seriously. They have a pre-formed set of “answers,” which appear to be based entirely on their own personal political preferences, and they present these whether or not they are even relevant to the question at hand.

C. Other Actions and Considerations

3. With respect to liquid nicotine and the dangers of nicotine poisoning, should FDA consider requiring any additional warnings beyond a nicotine exposure warning (text and/or any applicable color or graphic element)? If so, please describe the warning(s) (textual and/or graphic) and provide evidence or data to support your recommendation.

Yes. This issue was addressed in detail in our public comment on the FDA deeming rule, which is adapted below and submitted as a formal response to this question.
The FDA’s Proposed Warnings on Addiction are Inadequate and Do Not Reflect Current Understanding of Appropriate Messaging on Addiction (Page 25)

We will not respond to the remainder of the TCORS/CPCS document, which addresses the concept of addiction in relation to the products considered under this ANPRM. The authors have again ignored the actual question, “With respect to liquid nicotine and the dangers of nicotine poisoning . . . ,” and have taken this as an opportunity to introduce irrelevant material which has no bearing on acute poisoning and product handling safety.
In summary, the TCORS/CPCS comment, and by extension all other comments and commentators that take the same approach, is flawed to the point that the authors should not be considered reliable commentators on this or related matters. The comment is generally non-responsive to the questions posed by the ANPRM, and it is not clear the authors even understand the questions. The comment repeatedly invokes irrelevant, and in many cases, highly tendentious, false, or misleading material in support of specious reasoning, presenting this as if it were scientific analysis. The authors seem unaware of, and certainly unbothered by, the fact that many of their recommendations would significantly harm the public--in some cases, undermining the goals of the ANPRM--in exchange for little or no benefit.

In general, authors who demonstrate political bias, to the extent they bypass the questions at hand, and make recommendations consistent with their own policy goals without scientific support or analysis, represent a worrisome aspect of the FDA’s comment process. We trust the FDA will carefully scrutinize the TCORS/CPCS comment, and other comments that demonstrate such bias, and note the many instances where they depart from sound scientific practice in rulemaking and are antithetical to the purpose of the ANPRM. We further suggest that FDA generalize the knowledge gained from this scrutiny to recognize that other docket comments and published claims by like-minded individuals and organizations should be assumed to display similar bias and disdain for sound analysis, without the need for a point-by-point rebuttal of each as we provide here.




References

Albanese, M. J., & Shaffer, H. J. (2010). Addiction: the genesis and evolution of an idea and its consequences. In H. J. Shaffer(Ed.). APA Addiction Handbook (Vol. 1, pp. 3-25). Washington, DC: American Psychological Association.

Frenk, H., & Dar, R. (2000). A critique of nicotine addiction. Boston, MA: Kluwer Academic..

Frenk, H., & Dar, R. (2011). If the data contradict the theory, throw out the data: Nicotine addiction in the 2010 report of the Surgeon General. Harm Reduction Journal, 8:12.

No comments: